工作任务
语义模型相似度训练
学习任务
ABCNN 阅读和实现
学习收获
论文链接:ABCNN: Attention-Based Convolutional Neural Network for Modeling Sentence Pairs
关键词 :文本匹配 ,
answer selection (AS)给定一个问题,从候选答案集合中匹配最佳答案。
paraphrase identification (PI),给定两个句子,判断它们是否包含相同的语义。
textual entailment (TE) 给定一句话作为前提,另一句话作为推断,去判断能否根据前提得到推断。
entence completion (SC) 句子填空题(SC)
词性标签(pos_tag)
WikiQA dataset
(all-ap) For the output feature map of the last convolution layer, we do column-wise averaging over all columns, denoted as all-ap.This will generate a representation vector for each of the sentence,
(w-ap) For the output feature map of non-final convolution layers, we do column-wise averaging over windows of w consecutive columns, denoted as w-ap;